Procedia Computer Science 93 ( 2016 ) 453 – 461
Available online at www.sciencedirect.com
1877-0509 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the Organizing Committee of ICACC 2016
doi: 10.1016/j.procs.2016.07.233
ScienceDirect
6th International Conference On Advances In Computing & Communications, ICACC 2016, 6-8
September 2016, Cochin, India
Automatic Facial Expression Recognition Using DCNN
Veena Mayya, Radhika M. Pai
∗
, Manohara Pai M. M.
Department of Information & Communication Technology, Manipal Institute of Technology, Manipal University, Karnataka-576104, India
Abstract
Face depicts a wide range of information about identity, age, sex, race as well as emotional and mental state. Facial expressions
play crucial role in social interactions and commonly used in the behavioral interpretation of emotions. Automatic facial expression
recognition is one of the interesting and challenging problem in computer vision due to its potential applications such as Human
Computer Interaction(HCI), behavioral science, video games etc.
In this paper, a novel method for automatically recognizing facial expressions using Deep Convolutional Neural Network(DCNN)
features is proposed. The proposed model focuses on recognizing the facial expressions of an individual from a single image. The
feature extraction time is significantly reduced due to the usage of general purpose graphic processing unit ( GPGPU). From an
evaluation on two publicly available facial expression datasets, we have found that using DCNN features, we can achieve the
state-of-the-art recognition rate.
c
2016 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of the Organizing Committee of ICACC 2016.
Keywords:
Facial expression recognition;Computer Vision;Machine Learning;Confusion Matrix;Support vector machine; JAFFE; CK+; Deep
Convolutional Neural Network;
1. Introduction
Facial expression is an important part of nonverbal communication. Human expression recognition is influenced
by certain context. When a subject is being investigated, the investigator might be diverted by the subject’s voice
tone or argument and may forget to keep track of the facial expressions. Automatic facial expression recognition
systems are exempt to such contextual interference. Such systems can be beneficial in many fields, like gaming
applications, criminal interrogations, psychiatry, animations etc. State-of-art approaches attempt to recognize six
basic facial expressions such as anger, disgust happiness, sadness, surprise and fear.
Facial expression recognition techniques are based on either appearance features or geometry features
1
. Geometric
features are extracted from the shape of the face and its components such as the eyebrows, the mouth, the nose etc.
Appearance features are extracted using the texture of the face caused by expression, such as furrows, wrinkles etc.
In 1970s Paul Ekman and Wallace V. Friesen, developed Facial Action Coding System (FACS)
2
which is the most
∗
Corresponding author. Tel.: +91-08202925362;
E-mail address: radhika.pai@manipal.edu
© 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the Organizing Committee of ICACC 2016